Generation method and generation system of purchase technical specification book

Abstract

The invention discloses a method for generating a purchase technical specification book, which comprises the following steps of: acquiring a data set, classifying the data set, and giving corresponding keywords according to the content of the data set; selecting corresponding data from the data set according to a user demand, and generating an initial text according to a sequence; and verifying the initial text to finally obtain the technical specification template. By means of preset rules and algorithms, the technical specifications are automatically analyzed, problems can be quickly positioned and processed, the auditing time is greatly shortened, and the overall auditing efficiency is improved; an intelligent auditing technology is applied, a more accurate data field constraint is set, a data entry range is limited, deep logic verification is carried out, and data accuracy and normalization are ensured; according to existing standards and historical data, a specification book framework and content are automatically generated, reference is provided for compilation personnel, compiling formats are standardized, words and expressions are unified, and specialty and consistency of a specification book are improved.

Description

Technical Field

[0001] The present invention relates to a method for generating a specification document, in particular to a method for generating a procurement technical specification book for suppliers. The present invention also relates to a generation system using the method. Background Art

[0002] The traditional technical specification review process for authorized procurement by power companies relies primarily on manual review. Reviewers thoroughly review various technical specification documents, including extensive information such as construction location, completion date, and procurement method. Drawing on their understanding of power industry standards and procurement regulations, they scrutinize each document word-for-word to determine compliance. However, this method is highly inefficient, requiring significant time and effort from reviewers and prone to oversight. Consequently, some non-compliant technical specifications go unidentified, severely impacting procurement quality and corporate profitability. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for generating a procurement technical specification book, and the present invention also relates to a generation system using the method.

[0004] The present invention is implemented by the following technical means: a method for generating a procurement technical specification book, comprising the following steps: Obtain a data set, classify the data set, and assign corresponding keywords based on the content of the data set; According to user needs, select corresponding data from the data set and generate initial text in sequence; Verify the initial text and finally obtain the technical specification template; The data sets are classified according to service types, and the keywords include service types and the location of the data sets in the technical specifications.

[0005] The user requirements shall at least include the business scope, the organizing unit, and the relevant qualifications of the service provider.

[0006] The text verification includes the detection of typos and incorrect punctuation marks, the detection of extra spaces and paragraph marks, and the detection of inappropriate words.

[0007] The inappropriate word detection includes banned word detection, sensitive word detection, and discriminatory terms detection; The banned words include vague expressions, specific amounts and warranty periods; the sensitive words include guarantee or commitment words; and the discriminatory clauses include restrictions on specific suppliers and specific fields.

[0008] The specification book template is a doc document, and the content in the data set can be updated as needed.

[0009] A system for generating procurement technical specifications, comprising: Specification book generation module: used to generate technical specifications that comply with regulations; Input module, used to receive user input requirements and instructions for modifying the data set; Verification module, used to verify the content of the data set and the generated specification book; The storage module is used to store data sets and specification templates.

[0010] In the storage module, the data sets are classified according to the service type, and the keywords include the service type and the location of the data set in the technical specification book.

[0011] The user requirements received by the input module at least include business scope, organizing unit, and relevant qualifications of the business provider.

[0012] The verification module verifies the text by detecting misspellings and incorrect punctuation marks, extra spaces and paragraph marks, and inappropriate vocabulary.

[0013] The inappropriate word detection includes banned word detection, sensitive word detection, and discriminatory terms detection; The banned words include vague expressions, specific amounts and warranty periods; the sensitive words include guarantee or commitment words; and the discriminatory clauses include restrictions on specific suppliers and specific fields.

[0014] Compared with the existing technology, the present invention automatically analyzes technical specifications with the help of preset rules and algorithms, can quickly locate and handle problems, significantly reduce review time, and improve overall review efficiency; uses intelligent review technology to set more precise data field constraints, limit the scope of data entry, and perform in-depth logical verification to ensure data accuracy and standardization; based on existing standards and historical data, it automatically generates the framework and content of the specification, provides a reference for compilers, standardizes the writing format, unifies the wording and expression, and improves the professionalism and consistency of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a template management flow chart of the present invention; Figure 2 Generate a flow chart for the template of the present invention; Figure 3 Generate a flow chart for the technical specification of the present invention; Figure 4 Generate a flow chart for the quotation form of the present invention; Figure 5 This is the review flow chart of the technical specification of the present invention; Figure 6This is a diagram of the verification module review steps of the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings: The present invention relates to a method for generating a procurement technical specification, comprising the following steps: Obtain a data set, classify the data set, and assign corresponding keywords based on the content of the data set; According to user needs, select corresponding data from the data set and generate initial text in sequence; Verify the initial text and finally obtain the technical specification template; The data sets are classified according to service types, and the keywords include service types and the location of the data sets in the technical specifications.

[0017] The user requirements shall at least include the business scope, the organizing unit, and the relevant qualifications of the service provider.

[0018] The text verification includes the detection of typos and incorrect punctuation marks, the detection of extra spaces and paragraph marks, and the detection of inappropriate words.

[0019] The inappropriate word detection includes banned word detection, sensitive word detection, and discriminatory terms detection; The banned words include vague expressions, specific amounts and warranty periods; the sensitive words include guarantee or commitment words; and the discriminatory clauses include restrictions on specific suppliers and specific fields.

[0020] Specifically, text verification mainly examines three aspects: document structure, title and body. The document structure requires that the first-level directory and the second-level directory must be exactly the same as the following, otherwise they will be marked in yellow as a warning.

[0021] The title part requires that the "Project Name" in the technical specification is completely consistent with the "Tender Project Section Name" in the demand plan. The purchaser's abbreviation cannot contain any extra words or missing words, otherwise it will be marked in yellow as a warning.

[0022] Regarding format scanning of the main text, the review principle requires that the entire text must be free of typos, unnecessary spaces, and English punctuation marks. Otherwise, they will be marked in yellow as a warning. Regarding prohibited words, ambiguous expressions such as "regularly" and "not less than" are prohibited. Specific amounts and tax rates are prohibited. Words that contradict the construction period of the bid section are prohibited. Words similar to warranty periods, such as "provide a one-year quality guarantee," are prohibited. Regarding sensitive words, words such as "guarantee" and "commitment" are marked in yellow as a warning. Regarding errors in expression, warnings are issued for incorrect parts of the purchaser's company abbreviation, technical standards, and specifications. Regarding discriminatory clauses and supplier qualifications, specific suppliers or products cannot be targeted. Performance or awards in specific regions or industries cannot be used as bonus points or conditions for winning the bid. Specific patents, trademarks, or brands cannot be restricted or specified. The supplier's ownership form, organizational form, or location cannot be restricted. The supplier's scale and qualification requirements, such as registered capital and total assets, cannot be restricted. Regarding the quotation format, the unit price contracting method must indicate "quantity is tentatively estimated, settlement is based on actual results" and "the quantity is a tentative estimate, and the final settlement is based on the actual quantity incurred." Specific tables are not allowed.

[0023] The specification book template is a doc document, and the content in the data set can be updated as needed.

[0024] The present invention also relates to a system for generating procurement technical specifications, comprising: Specification book generation module: used to generate technical specifications that comply with regulations; Input module, used to receive user input requirements and instructions for modifying the data set; Verification module, used to verify the content of the data set and the generated specification book; The storage module is used to store data sets and specification templates.

[0025] In the storage module, the data sets are classified according to the service type, and the keywords include the service type and the location of the data set in the technical specification book.

[0026] The user requirements received by the input module at least include business scope, organizing unit, and relevant qualifications of the business provider.

[0027] The verification module verifies the text by detecting misspellings and incorrect punctuation marks, extra spaces and paragraph marks, and inappropriate vocabulary.

[0028] The inappropriate word detection includes banned word detection, sensitive word detection, and discriminatory terms detection; The banned words include vague expressions, specific amounts and warranty periods; the sensitive words include guarantee or commitment words; and the discriminatory clauses include restrictions on specific suppliers and specific fields.

[0029] The generation system supports users to add or delete management operations on the technical specification templates they upload. Specifically, after entering the interface, the currently existing template will be loaded by default; click the Browse button, select a template information, enter the template name, and then click the Add button to automatically read the template information and add it to the template list; select a template information, it will be filled into the template path and name above, click the Delete button to delete it; click a template information, the token list on the right will display the token information retrieved by the template, if there is any missing or redundant, you need to check the template and add it again; click the Save button to save all changes, and the saved information will be loaded next time.

[0030] The generation system of the present invention supports users to select locally existing templates, generate an input information interface, and complete the general specification book generation function. Specifically, all saved template name information will be displayed in the template list, and one of the templates can be selected; click the generation interface, and the corresponding input box will be generated according to the selected template. If there is a basic data table, a list will be generated for inputting information; click the browse button to open a directory selection box and select the directory where the generated word is stored; enter the file name; fill in the input box information after the interface is generated, click the start generation button, and the word will be generated according to the current information; click the reset button to clear the generated text box and file name information, which is convenient for generating the next file.

[0031] Based on user input and the selection of a specific technical specification template, the fields in the template can be replaced with the specified input information, allowing for quick output of the technical specification. Specifically, select a directory to store the generated Word document; then select the path to the technical specification template; enter information such as the file name and company name; if there are technical specifications and standard supplements, enter them one by one, with each record representing one piece of supplementary information; click Start Generation and wait for the Word document to pop up. If no modifications are required, simply close it after review; click the Reset button to clear the file name and the following content, re-enter the information, and continue generating the next Word document.

[0032] The generation system of the present invention can also generate a quotation sheet in a specified format based on the input information. Specifically, first select the storage directory for the generated quotation sheet; enter the required information such as the project name and sub-item name; click Add, and a new record will be added to the table below, indicating that there is a sub-item in the quotation sheet; click Delete, select a record in the list, and click the Delete button to delete the selected record; after clicking Add, the sub-item name and other information will be cleared, and you can continue to enter and add, or click Start Generation; wait for the Excel document to pop up, modify or directly close it, and the generation is successful; click Reset, which will clear the project name and all the information below, and you can continue to generate the next one.

[0033] The present invention can also review the technical specification book after the user has completed filling it out. Specifically, first select the directory where the document to be reviewed is located; then select the location of the demand plan table; click to start the review and wait for the result pop-up window to complete the review; after the review is completed, the review document result table will have the status of all documents, success or failure; select one of the records, such as a failed record, and all error details of the document will be displayed in the problem details table on the left, including page numbers, line numbers and other information.

[0034] The scope of review of the procurement technical specifications of power enterprises includes the following: 1. Completeness review. Ensure that the document is complete and covers key clauses. Check that the document includes essential sections such as technical parameters, acceptance criteria, scope of supply, and after-sales service. Ensure that the specification covers mandatory industry clauses such as safety standards, environmental requirements, and energy efficiency ratings. Ensure that the document is complete and meets industry requirements.

[0035] 2. Compliance Review. This audit requires compliance with regulatory standards and internal control requirements. The technical parameters and test methods in the document are compared with current national / industry standards such as GB, DL, and IEC to verify compliance with regulatory standards. The audit also checks whether the internal control requirements of the company's procurement policy regarding supplier qualifications and payment terms are met.

[0036] 3. Technical Review: This requires that parameters be logically sound and terminology standardized. This review verifies the logic between technical parameters, ensuring that parameters such as voltage level and equipment capacity match each other. It also identifies non-standard terminology or ambiguous expressions such as "approximately" and "in principle" in the document to ensure the standardization of technical content.

[0037] 4. Consistency review. This requires consistency across chapters and text and graphics. Check whether technical parameters, acceptance criteria, and other content are consistently expressed in different sections of the document. Also, proofread technical drawings and text descriptions to ensure that device dimensions, interface types, and other graphic content match correctly.

[0038] 5. Risk Review: This requires accurate identification of ambiguous clauses and clear division of responsibilities. This includes identifying ambiguous clauses in documents that could potentially lead to disputes, such as non-quantifiable acceptance criteria. Furthermore, it requires clarifying the supplier's responsibilities regarding warranty periods, breach of contract, and other areas to mitigate potential risks.

[0039] 6. Multimodal content review. This requires reviewing graphic content and processing unstructured text. Image and text recognition technology (such as YOLO + OCR) is used to review images and charts in documents to ensure they contain necessary captions and are consistent with the technical description. Natural language processing technology is also used to process unstructured text to ensure that key information is not missed.

Claims

1. A method for generating a procurement technical specification, characterized in that: The steps include: Obtain a data set, classify the data set, and assign corresponding keywords based on the content of the data set; According to user needs, select corresponding data from the data set and generate initial text in sequence; The initial text is verified and finally a technical specification template is obtained.

2. The method for generating a procurement technical specification according to claim 1, characterized in that: The data sets are classified according to service types, and the keywords include service types and the location of the data sets in the technical specifications.

3. The method for generating a procurement technical specification according to claim 1, characterized in that: The user requirements shall at least include the business scope, the organizing unit, and the relevant qualifications of the service provider.

4. The method for generating a procurement technical specification according to claim 1, characterized in that: The text verification includes the detection of typos and incorrect punctuation marks, the detection of extra spaces and paragraph marks, and the detection of inappropriate words.

5. The method for generating a procurement technical specification according to claim 4, characterized in that: The inappropriate word detection includes banned word detection, sensitive word detection, and discriminatory terms detection; The banned words include vague expressions, specific amounts and warranty periods; the sensitive words include guarantee or commitment words; and the discriminatory clauses include restrictions on specific suppliers and specific fields.

6. The method for generating a procurement technical specification according to claim 1, characterized in that: The specification book template is a doc document, and the content in the data set can be updated as needed.

7. A system for generating procurement technical specifications, characterized in that: include: Specification book generation module: used to generate technical specifications that comply with regulations; Input module, used to receive user input requirements and instructions for modifying the data set; Verification module, used to verify the content of the data set and the generated specification book; The storage module is used to store data sets and specification templates.

8. The system for generating procurement technical specifications according to claim 7, characterized in that: In the storage module, the data sets are classified according to the service type, and the keywords include the service type and the location of the data set in the technical specification book.

9. The system for generating procurement technical specifications according to claim 7, characterized in that: The user requirements received by the input module at least include business scope, organizing unit, and relevant qualifications of the business provider.

10. The system for generating procurement technical specifications according to claim 7, characterized in that: The verification module verifies the text by detecting misspellings and incorrect punctuation marks, extra spaces and paragraph marks, and inappropriate vocabulary.

11. The system for generating procurement technical specifications according to claim 7, characterized in that: The inappropriate word detection includes banned word detection, sensitive word detection, and discriminatory terms detection; The banned words include vague expressions, specific amounts and warranty periods; the sensitive words include guarantee or commitment words; and the discriminatory clauses include restrictions on specific suppliers and specific fields.

Images